The delivery of drug species to the brain and other organs is often seriously limited by transport and metabolism factors, including biological membranes; specifically, in the case of the brain, delivery is limited by the functional barrier of the endothelial brain capillary wall, i.e. the blood-brain barrier or BBB. Site-specific and sustained delivery of drugs to the brain or other organs, i.e. targeted drug delivery, is even more difficult.
Many drugs are hydrophilic and are unable to penetrate the brain to any considerable extent. Other drugs which are lipophilic and/or for which particular transport mechanisms exist may be able to cross the BBB and reach the brain, but the very lipophilicity which enables their entry likewise facilitates their exit. It is thus necessary to administer large doses of drugs to achieve adequate brain levels (if, indeed, such is even possible), and this in turn overburdens non-targeted loci and results in significant toxicity.
It is now well-known that numerous drugs exert their biological effects through centrally-mediated mechanisms. Thus, a brain-targeted approach is a desirable means of delivery for a wide diversity of drugs, including neurotransmitters, stimulants, dopaminergic agents, tranquilizers, antidepressants, narcotic analgesics, narcotic antagonists, sedatives, hypnotics, anesthetics, antiepileptics/anticonvulsants, hormones such as the male and female sex hormones, peptides, anti-inflammatory steroids, non-steroidal anti-inflammatory agents/non-narcotic analgesics, memory enhancers, antibacterials/antibiotics, antineoplastics (anticancer/antitumor agents) and antiviral agents.
In recent years, the need for more effective treatment of a number of viral disease states has become increasingly urgent. The generally poor therapeutic accessibility of viral infections can be traced to three major facets including the viral life cycle, the lack of efficacious pharmacologically-active agents and, finally, the inability to deliver those agents which are available to the central nervous system (CNS) for sustained periods and in significant amounts.
Viruses are submicroscopic pathogens which depend on the cellular nucleic acid and protein synthesizing mechanisms of its host for propagation. In general, viruses invade cells by first interacting at a recognizable surface protein, penetrating the cell membrane and subsequently releasing themselves from a protective polypeptide coat to eject the core of the virus. The heart of these pathogens is genetic material, either DNA or RNA, and the type of nucleic acid gives rise to the system of nomenclature for these entities. The viral DNA and RNA can interact with cellular components to produce daughter genetic material as well as various structural or enzymatic proteins. After assembly and release, the viral progeny may infect other cells, yielding disease or ultimately death.
DNA viruses are subdivided into five families and include the pathogens responsible for labial and genital herpes, herpes encephalitis, human cytomegalovirus infection, chicken pox, shingles and mononucleosis. RNA viruses are present in more numerous forms and are subdivided into ten families. These viruses are unusual in that they reverse the usual DNA.fwdarw.RNA.fwdarw.protein sequence which occurs in higher life forms. RNA viruses are unusually dangerous for several reasons, including their lethality and the lack of effective treatments. RNA viral diseases include acquired immune deficiency syndrome, hemorrhagic fevers of various descriptions, Dengue fever, Lassa fever, and numerous encephalitic maladies including Japanese B encephalitis.
Chemotherapeutically, very few antiviral agents have been developed that have high in vitro activity against these viruses. One notable advance in the field was the advent of ribavirin or 1-.beta.-D-ribofuranosyl-1,2,4-triazole-3-carboxamide, synthesized in 1972. Ribavirin has a broad range of activity against both DNA and RNA viruses. This riboside, which contains an unnatural triazole base, significantly suppresses the infectivity and cytopathicity of several viral pathogens by mechanisms which are as of yet unclear. Several interactions have been suggested including inhibition of viral RNA polymerase, the inhibition of inosine monophosphate dehydrogenase by ribavirin anabolites and interference of mRNA cap formation by the 5'-triphosphate of ribavirin.
Ribavirin is active against several influenza viruses and respiratory syncytial virus and as such is used in an aerosol form to treat these diseases. Ribavirin is also used in the treatment of Lassa fever which rages in epidemic proportions in Sierra Leone. Unfortunately, while peripheral viral infections can be successfully treated with ribavirin and other riboside derivatives, encephalitis is immune to the action of these drugs. The inability of antiviral drugs, which are highly potent in vitro, to exert activity in the CNS is attributable to their exclusion from the brain. The basis of this impermeability is the blood-brain barrier (BBB), which effectively separates the systemic circulation from the brain parenchyma. As this barrier is lipoidal in nature, the BBB restricts the entry of materials which do not have high affinity for the phospholipid matrix and consequently hydrophilic compounds are excluded. Thus, drug molecules must be intrinsically lipophilic if they are to gain access to the CNS. This is the restriction which renders ribavirin, which has a log P of only 2.06, ineffective in treating viral diseases of the brain.
Many antiherpetic agents exhibit poor penetration across biological barriers such as the BBB and the ocular and skin barriers, achieving concentrations well below therapeutic levels. Improved delivery of an antiherpetic agent across these barriers would offer a significant advantage in the treatment of such serious and debilitating diseases as encephalitis, ophthalmic infections caused by herpes simplex such as herpetic uveites, keratitis etc. and cutaneous herpes infections such as genital and orofacial herpes.
Vidarabine (9-.beta.-D-arabinofuranosyladenine, Ara-A, adenine arabinoside) is a purine nucleoside analog with a broad spectrum of antiviral activity against a number of DNA viruses, including HSV-1 and 2, cytomegalovirus and varicella zoster virus. The drug has been shown useful in the treatment of brain biopsy-proven herpes simplex encephalitis (HSE), resulting in a statistically significant reduction in mortality. Ara-A has demonstrated clinical utility as a topical agent for herpes keratitis of the eye. However, when applied locally to the skin, vidarabine has provided no benefit in genital or orafacial HSV infection. In immunocompromised patients with localized herpes zoster, Ara-A has demonstrated a beneficial effect in accelerating cutaneous healing and decreasing the rate of cutaneous dissemination.
The essential mechanism of inhibition of viral replication by vidarabine, although not precisely defined, appears to be a consequence of the incorporation of the drug into viral DNA. To exert its antiviral action, vidarabine must fast be phosphorylated by cellular enzymes to the triphosphate, which competitively inhibits HSV DNA polymerase. Some investigators have found that the viral DNA polymerase activity is more sensitive to inhibition than that of cellular DNA polymerases, an observation that could explain some of the selective toxicity of the drug and its dose-related toxicity. Vidarabine triphosphate is incorporated into both cellular and viral DNA, where it may act as a chain terminator for newly synthesized HSV nucleic acid.
Despite its proven efficacy, Ara-A does suffer from a number of limitations, including low lipophilicity as evidenced by a negative log P (octanol/water), which results in a failure to be adequately transported across biological membranes.
Herpes simplex virus is a causative factor for encephalitis. Its involvement in the CNS represents the most common cause of nonepidemic fatal encephalitis in the United States. An estimated 1,000 to 5,000 cases occur each year in the U.S., with death in over one half of those who are untreated. Herpes simplex virus type 2 causes encephalitis in patients with thymic dyplasia and other severe immunodeficiency states. Encephalitis also is a common opportunistic infection associated with AIDS.
The acute severe encephalitis due to herpes simplex type 1 in humans may represent a primary infection, a reinfection or an activation of latent infection. The primary mode of viral transport into the CNS has not been clearly established. However, it has been shown that following extraneural inoculation, the virus gained access to the CNS by both hematogenous and neural pathways. The neural pathway of transport in man is supported by the fact that the virus can be isolated from explants of both trigeminal ganglia in the majority of routine autopsies.
Herpes simplex encephalitis is the most common cause of sporadic fatal encephalitis. Both the high mortality rate and the risk of severe sequelae in the survivor have prompted attempts at therapy with antiviral compounds. In order for the antiencephalitic agent to exert its effect, it is necessary for the drug to be present in the CNS where the virus is lodged, at an optimum concentration and for a sufficient period of time. Maintaining a therapeutic level of the drug over a prolonged period at the site of action is essential in optimal reduction of viral concentrations. Resistance of virus in the brain after treatment has been reported in almost all of the cases studied so far. Only very rarely has total remission been achieved.
The main reason for the lack of successful treatment is the inefficient method of drug delivery to the brain, the major impediment to drug delivery to the brain being the blood-brain barrier. Antiviral agents such as iododeoxyuridine and vidarabine exhibit little activity and high toxicity in the treatment of encephalitis. This is primarily due to their inability to cross the blood-brain barrier at optimum concentrations. In the case of other antivirals such as acyclovir, drug resistance has been observed. To overcome such problems, a new family of fluorinated nucleoside analogs has been synthesized. This family includes 1-(2'-deoxy-2'-fluoro-.beta.-D-arabinofuranosyl) derivatives of 5-methyluracil (FMAU), 5-iodocytosine (FIAC) and 5-iodouracil (FIAU). FIAU is a metabolite of FIAC. These compounds have been shown to display significant antiviral activity against herpes viruses in vitro and in some in vivo experiments. The mechanism of antiviral activity depends in part on the phosphorylation of these agents by viral-specified thymidine kinase. These agents are rapidly taken up and phosphorylated only to the 5'-monophosphate in HSV-infected cells; the monophosphates are presumably further phosphorylated by cellular enzymes to the corresponding triphosphates. Phosphorylation of these agents by the virus-coded thymidine kinase is much better than by the cellular enzymes. These antiviral agents are incorporated into termini and internucleoside linkages of viral DNA much more than into the DNA of uninfected cells. Since maximum selectivity would improve the therapeutic potential of any new antiviral drug, relatively low toxicity with normal cells is mandatory. The low cytotoxicity exhibited by these agents with uninfected cells indicate selectivity of action.
Although these nucleoside analogs exhibit high selectivity toward viral cells, they are quite polar and therefore their ability to penetrate the BBB is greatly minimized. They must be administered in high doses to attain an effective level in the brain, resulting in severely toxic side-effects. For example, FMAU, considered the most potent antiviral agent of its class (therapeutic index greater than 3,000) in treating encephalitis, produces irreversible neurological damage at doses greater than 32 mg; other side effects include diarrhea, nausea and blood count depression. High doses of FIAU have resulted in cardiac fibrosis, myelosuppression and lymphoid depletion. In the case of FIAC and FMAU, significant reduction in body weight or death has also been noted at higher doses. Further, sustained therapeutic levels have not been achieved, even at these higher doses.
It is known that FIAC is metabolized extensively in vivo and that its metabolites retain their antiviral activity in cell culture. The major metabolites of FIAC include the deaminated species FIAU, the deiodinated species 2'-fluoroarbinosylcytosine (FAC) and 2'-fluoroarabinosyluracil (FAU) and their glucuronides. Two metabolites of FMAU have been isolated from the urine of mice. These include 2'-fluoro-5-hydroxymethylarabinosyluracil (FHMAU) and a glucuronide of FMAU. FMAU, FIAU and FIAC have been found to exhibit more potent antiviral activity than acyclovir. The metabolites of these compounds, even though potent inhibitors of HSV-2 in cell cultures, are essentially devoid of antiviral activity in vivo in the encephalitis model. This dichotomy between in vitro activity and in vivo activity suggests that these agents do not cross the BBB in sufficient concentration to exert activity.
(E)-5-(2-bromovinyl)deoxyuridine (BVDU) is also a polar antiviral agent effective against encephalitis caused by herpes zoster virus and HSV-1. This agent crosses the BBB in low levels only at very high concentrations; as a result, it has been shown to induce sister chromatid exchange. Other side-effects include toxicity to liver, bone marrow function and gonads.
Dihydroxypropoxymethylgaunine (DHPG) belongs to the same class of antiviral agents as acyclovir. However, DHPG has been shown to be at least 100-fold more effective than acyclovir in the treatment of encephalitis in vitro and in vivo. DHPG is more efficiently phosphorylated in infected cells than is acyclovir. As with acyclovir, herpes virus-specific thymidine kinase phosphorylates DHPG to its monophosphate, which is further phosphorylated to its di- and triphosphate by cellular guanylate kinase and other cellular enzymes, respectively. However, DHPG is transported to the brain only at high doses, which in turn produce high plasma levels of the drug which exert cytotoxic effects on normal human mycloid cells. Studies have shown that acyclovir crosses the BBB poorly, and at higher doses causes problems such as renal blockage.
Human cytomegalovirus (HCMV) is a virus of the herpes group which includes herpes simplex I and II, Epstein-Barr virus, and varicella zoster virus. In common with the other members of its group, infection with HCMV leads to a latent state in which the viral genome becomes incorporated in the host DNA, and in which recurrent infections are common. Vital infection with HCMV is quite widespread, with approximately 50% of Americans showing seropositivity by age 30. In the majority of cases the virus does not cause an overt disease state, but can be detected through serological and other laboratory procedures in otherwise healthy individuals. In the absence of complicating factors, exposure to the virus can result in a clinical presentation ranging from asymptomatic seroconversion to a disease state resembling infectious mononucleosis.
In contrast to viral infection in normal adults, HCMV in the fetus or neonate can result in severe clinical manifestations. The virus in these cases is acquired congenitally, often from asymptomatic mothers. The virus has been said to be the single most frequent cause of viral infections in newborns. The occurrence of HCMV in neonates is from 0.5% to 4% of all live births, but only 10% to 20% of these will have clinical manifestations of cytomegalic disease, which mainly involve the CNS and which can result in permanent, debilitating brain damage or auditory degeneration.
When the host immune system is suppressed, HCMV becomes a much more serious infective agent. In this state, a latent HCMV infection may recur, or a primary infection may be unusually severe. Immunosuppression can occur in several circumstances, for example, during use of immunosuppressive drugs, such as corticosteroids, azathioprine, and thymocyte immune globulin which are given to prevent rejection of a transplanted organ when a patient has undergone organ transplant surgery. Along with other complications, cytomegalic disease is a common and sometimes especially serious problem which can follow successful kidney, bone marrow, and heart transplantation. The manifestations of cytomegalic disease following transplant surgery can include, but are not limited to, retinitis and pneumonitis. Another particularly serious complication occurring during immunosuppressive therapy is Kaposi's sarcoma (KS). A strong correlation is known to exist between KS and HCMV, to the extent that it has been postulated that HCMV causes KS, analogously to the relationship between Epstein-Barr virus and Burkitt's lymphoma. However, a causal role for the virus has not been definitively established.
An immunosuppressed state is the hallmark of acquired immunodeficiency syndrome (AIDS), and HCMV has been shown to have an extraordinary prevalence in this population, approaching 94%. In addition, cytomegalic disease and its complications are among the primary causes of much of the suffering from AIDS as well as a major factor causing death. HCMV is known to result in a suppression of cell-mediated immunity through depression of levels of T-helper cells with an increase in suppressor/cytotoxic T-cells. Before the discovery of human immunodeficiency virus (HIV), the list of candidates for the cause of AIDS included HCMV. The consequences of HCMV infection in AIDS are manifold, with neural and especially ocular involvement being predominant. Ocular involvement is presented as a hemorrhagic retinitis, first evidenced by blurring of vision. This retinitis is so common that it has been proposed that it be the primary diagnostic evidence for the presence of AIDS. Neural involvement resulting in viral encephalitis is also common and presents itself post-mortem in the microglial nodules which are typical of HCMV infection. In AIDS, this neural involvement is concomitant with HIV infection of the CNS, often manifesting as subacute encephalopathy.
An antiviral agent which has shown promise in the treatment of HCMV infections in immunosuppressed states is DHPG. As mentioned above, DHPG is structurally similar to acyclovir (ACV), a safe and efficacious antiherpetic agent. The primary mechanism of DHPG action against CMV is inhibition of the replication of viral DNA by DHPG-triphosphate. This inhibition includes a selective and potent inhibition of the viral DNA polymerase. Since HCMV does not encode a virus-specific thymidine kinase, phosphorylation of DHPG is presumably accomplished by the host-cell enzymes, primarily various nucleoside kinases, which have been shown to be elevated in HCMV-infected cells. The markedly increased activity of DHPG toward CMV compared with ACV appears to be due in part to the efficient intracellular metabolism of DHPG to its mono and triphosphate in CMV-infected cells. The relative in vitro activities, as measured by the IC.sub.50 values of DHPG vs ACV are of the same order against herpes simplex virus (HSV), namely 0.2 to 0.8 .mu.M. However, against HCMV the IC.sub.50 for DHPG is approximately 2.5 .mu.M. Thus, DHPG has significant activity against HCMV in vitro. These promising results have been extended in animal models as well as in clinical trials.
As mentioned above, one of the rust clinical signs of AIDS infection is a retinitis which is caused by HCMV. One of the most dramatic recent clinical demonstrations of antiviral activity has been in a study of the effects of intravenous DHPG in AIDS patients who were suffering from progressive blindness caused by cytomegalic infection of the retina. In these patients, not only did viral liters drop to an unobservable level, but clinically observable improvement in sight was achieved. In other studies, significant improvement in other areas of cytomegalic infection was shown. These included improvement in the cytomegalic pneumonitis and encephalitis, as well as gastrointestinal infections.
DHPG, obviously, has very high intrinsic activity but, as with most useful drugs, has a number of inherent undesirable properties as well. Problems with the aqueous solubility of the compound (5.1 mg/mL at 37.degree. C.) necessitate the use of the sodium salt for the intravenous administration of the drug. This induces pain or phlebitis at the infusion site, since the pH of the solution is about 11. In humans, oral bioavailability of DHPG is only 3-4.6% based on urinary excretion, with 99% of the drug being excreted unchanged by the kidneys. The pharmacokinetic disposition of intravenous DHPG in humans is similar to that observed in rats and dogs, with the finding of a biphasic elimination with an .alpha.-phase half-life of 0.23 hours and a .beta.-phase of 2.53 hours. These values are quite similar to those for acyclovir, and show that repeated dosing is necessary to maintain effective plasma concentration. Neutropenia is the most frequent dose-dependent toxicity associated with DHPG therapy.
DHPG is a hydroxymethyl analog of acyclovir and consequently is more polar and is expected to pass through the blood brain barrier (BBB) even less readily. In rodent models, it has been shown that acyclovir distributes into most organs, with the highest levels found in renal tissue and the lowest levels found in brain tissue. Pharmacokinetic studies of DHPG in the rat and dog have demonstrated behaviour similar to acyclovir. Human pharmacokinetics of intravenous DHPG indicate cerebrospinal fluid (CSF) concentrations equivalent to 24% to 67% of plasma concentrations. However, since CSF levels may reflect transport through the choroid plexus, some uncertainty regarding specific brain levels of DHPG exists. Regardless of the efficiency with which DHPG crosses the BBB, however, it is to be expected that it may leave the CNS by the same mechanism with equal facility. In view of the significant role played by CMV in AIDS patients with severe neurologic complications, and the possibility that CMV could create a reservoir of persistent infection of the CNS even if peripheral clearance were realized, there exists a rationale for identifying antiviral drugs that can penetrate the BBB and accumulate in the brain, thereby providing a sustained release of the antiviral to maintain a therapeutically effective concentration.
Acquired immune deficiency syndrome (AIDS) was first described as a distinct clinical entity in 1981. As of October 1989, 110,000 cases of AIDS, as defined by the Center for Disease Control (CDC), have been diagnosed and 65,000 people have died from the disease. This insidious and pernicious malady has a 2-3 year fatality rate of almost 100% and is expected to strike between 135,000 and 270,000 people by 1991 alone. AIDS is now the leading cause of premature mortality in a number of areas and in several subpopulations in the U.S.; by 1991, it is expected to be a major killer. In other areas of the world, a similarly grim picture is developing. In central Africa, where the AIDS pathogen evolved, the disease is endemic and in several locations the increase in incidence of infection exceeds 0.75% of the total population per year. AIDS is caused by a retrovirus related to the lentivirinae family and has been designated human immunodeficiency virus (HIV-1). This pathogen selectively infects lymphocytes bearing a T4 surface antigen. These helper/inducer T-cells are responsible for containing and eliminating various types of infection including those precipitated by Pneumocystis carinii, Toxoplasma gondii, Cryptococcus neoformans, Candida albicans, Mycobacterium aviumintracellular and others. The destruction of cellular immunity induced by HIV-1 causes the normally benign infections resulting from the above-mentioned pathogens to run more fulminate courses. These opportunistic infections are generally the causes of death in patients with AIDS.
Early in the course of the AIDS epidemic, clinicians noted that patients were depressed and initially this was attributed to a normal psychological response to learning that one had a terminal disease. Later, however, it was realized that cognitive impairment and dementia were associated with AIDS. These CNS-associated symptoms of AIDS are now well-recognized and affect 40% of all AIDS patients at some point in the course of the disease.
In AIDS, the CNS, like the periphery, is susceptible to opportunistic infections and unusual neoplasms. Several of these have been identified, including cerebral toxoplasmosis, cryptococcal infection, candidiasis, cerebral tuberculosis, progressive multifocal leukoencephalopathy, cytomegalovirus encephalitis and primary brain lymphomas. Interestingly, these occur in less than 30% of neurologically-impaired AIDS patients. In addition, symptoms caused by these pathogens are generally focal in nature and are expressed as seizures. In the majority of AIDS patients, neuropsychiatric changes are characterized as an insidious, progressive dementia related to diffuse parenchymal brain dysfunction. Early symptoms of this disease include impaired cognitive, motor and behavior functions, including the inability to concentrate, difficulty in recalling recent events, losing one's train of thought in midsentence and general mental slowing. Motor impairments include leg weakness and problems in proprioception. Behaviorally, victims become apathetic, withdrawn and distraught. Later symptoms include global cognitive dysfunction with psychomotor retardation. Victims are autistic, mute, lethargic and quietly confused. Patients manifest urinary and fecal incontinence and may be afflicted by painful peripheral neuropathies including burning sensations or numbness. Neurohistopathologically, the picture is even worse. While only 40% of AIDS patients are recognized as demonstrating brain dysfunction, 80-95% of the brains from AIDS patients are abnormal at autopsy. Gross changes include decreased brain weight and general cerebral atrophy. Histopathologically, several unique abnormalities are consistently seen in demented AIDS patients. Most of these are white matter changes and include a diffuse pallor, perivascular and parenchymal sites that contain lymphocytic and macrophage infiltrates and vacuolation. Other changes include the presence of microglial nodules whic infect both gray and white matter and bizarre giant multinucleated cells. The presence and number of these cells which contain HIV-1 virons give excellent correlation with the severity of the dementia. The agent responsible for subacute encephalitis, also known as AIDS encephalopathy, has been shown to be HIV-1. Several direct and indirect lines of evidence support this etiology.
This central infection will have a detrimental impact on possible therapies directed at AIDS. The CNS is protected by the BBB and is not drained by the lymphatic system, making it an excellent location for eluding the immune system. If, therefore, agents are found that reconstitute the immune system, peripheral manifestation of AIDS, including many opportunistic infections, can be cured but the central infection will persist. The result of this could be a physically healthy but severely demented individual. In addition, host-cell restriction, i.e. partial expression of the viral genome, may cause viral latency in the CNS for many years. Also, once proviral DNA is incorporated, the only hope of containing the disease is by preventing the spread of further cellular infection. This implies, based on active in vitro doses, that for antiviral therapies to be effective, agents must pass the BBB and achieve relatively high sustained levels in neural tissue. The neurotropic nature of HIV-1 and the fact that the brain probably acts as a viral reservoir makes implementing the preceding statement imperative. Of agents presently available, azidothymidine (also known as zidovudine or AZT) has been clinically shown to be the most useful in mitigating the effects of the AIDS virus. AZT inhibits retroviral transcriptase, the enzyme responsible for initiating viral replication.
AZT has been shown to improve the immunological picture in AIDS patients. In various clinical studies, T-cell lymphocytes (T4.sup.+) were shown to increase in number, opportunistic infections spontaneously disappeared, and patients gained weight due to increased appetite. Also, fever subsided and skin hypersensitivity returned. At high doses of AZT, viremia disappeared and T-cell function was restored. The bioavailability is about 60%. The drug is generally well-tolerated, but several untoward side effects occurred, including headache and abdominal discomfort. The most severe side effect was anemia, which proved to be dose-limiting in several cases. AZT has been used in large clinical trials, the results of which are very exciting. In a double blind study, 16 out of 137 died in the placebo group while only one patient out of 145 died in the AZT treatment group (250 mg/4 hrs). T4.sup.+ lymphocytes were higher in the treated group and fewer opportunistic infections occurred. As before, a reversible bone marrow depression resulting in granulocytopenia, thrombocytopenia, etc., was observed. Recently, dideoxyinosine has also been shown to be effective in reducing the cytopathicity and infectivity of HIV in vivo. The effect of AZT on the neurological manifestation of AIDS has been reported by Yarchoan et at, Lancet, i, 132 (1987). In a series of four case reports, AZT was shown to improve immunological and neurologic functioning. T4.sup.+ cells increased in number, motor symptoms improved, gait became less ataxic and muscle strength returned. Attention span increased in one case and verbal skills improved. Unfortunately, when the drug was stopped in cases of anemia, all improvements disappeared and mental function declined. This initial report indicated that AZT can at least partially reverse neurological dysfunction. The authors noted at the end of the paper that "even modest enhancement of BBB penetration might have very important clinical consequences".
These limited improvements in neurological symptomatology are consistent with the similarly limited ability of AZT to pass into the CSF. Unfortunately, CSF levels of a drug may be a poor indication of brain tissue levels. Several studies have shown that the correlation between CSF and parenchyma concentrations are not necessarily significant. In general, polar compounds such as AZT are the most deceptive in this respect. The reason for this is that if a hydrophilic compound is taken up primarily via an unprotected area like the choroid plexus, detectable concentrations may indeed reach the CSF but the compound may not partition into the lipoidal brain parenchyma and as a result may be restricted to the CSF. This would be manifested by apparently adequate AZT levels as measured by CSF sampling but inadequate levels in brain tissue where the drug is needed. This assumption has been borne out in a recent paper by Terasaki et al, J. Inf. Dis., 158, 630 (1988). In it, the BBB penetration of AZT was shown to be very low, close to the uptake of sucrose, a vascular marker. The high concentrations of AZT found in CSF are presumably due to active transport of AZT at the choroid plexus via the thymidine pump. Again, these CSF levels represent AZT which is not in equilibrium with the brain interstitial fluid and therefore is not accessible to infected sites. It is clear that high levels of AZT are required to provide even marginal improvement in AIDS encephalopathy and that these doses are peripherally toxic.
The previous discussion has indicated that the AIDS virus is neurotropic and that the resulting brain infection by this pathogen is disastrous. Various agents have been identified which inhibit infection and abolish cytopathology in the AIDS virus. In some instances these compounds, like AZT, pass the BBB and achieve quantitative levels in CSF. Clinical results suggest, however, that high sustained drug levels, i.e. those that approach in vitro inhibitory concentrations, are required in the brain. Importantly, CSF levels do not reflect brain tissue concentration of AZT. Unfortunately, simply increasing the dose proportionally to achieve these ends increases blood concentrations and leads to various dose-related toxicities. Anemia has proved to be dose-limiting in many cases with AZT. Increasing brain levels of the nucleoside is possible by administering lipophilic esters of AZT leading to an increase in brain concentration of the nucleoside. These prodrugs are, however, not optimized in terms of pharmacokinetics and tissue distribution. Thus, while it is true that by increasing the lipophilicity of AZT, the drug will more easily pass the BBB and enter the CNS, the increased lipophilicity will increase the distribution of the compound in general, leading to an even greater tissue burden in all locations. This has ramifications in terms of peripheral toxicity such as anemia, i.e. a bad situation is made even worse. The other major drawback of simply increasing the lipophilicity of AZT is that while influx to the CNS is increased, the efflux is also greater, with the result being poor retention in the CNS and a therapeutically insufficient biological half-life. These two objections to simple antiviral prodrugs, namely: 1) increased tissue burden with little tissue specificity, and 2) poor CNS retention, point to the need for a more sophisticated approach, i.e. a chemical delivery system for brain-targeted drug delivery.
A dihydropyridine.revreaction.pyridinium salt redox carrier system has recently been successfully applied to brain-targeted delivery of a variety of drug species. Generally speaking, according to that system, a dihydropyridine carrier moiety is covalently bonded to a biologically active compound, which derivative can enter the CNS through the blood-brain barrier following its systemic administration. Subsequent oxidation of the dihydropyridine species to the corresponding pyridinium salt leads to delivery of the drug to the brain.
More specifically, the redox carrier system provides for brain-targeted drug delivery by means of carrier-drugs, which in their reduced form, which is the form intended for administration, can be represented by the formula EQU [D-DHC]
wherein [D] is a centrally acting drug species and [DHC] is the reduced, biooxidizable, blood-brain barrier penetrating, lipoidal form of a dihydropyridine.revreaction.pyridinium salt redox carrier. In their oxidized form, which is the form "locked" in the brain from which the active drug is ultimately released, the carrier-drugs can be represented by the formula EQU [D-QC].sup.+ X.sup.-
wherein X.sup.- is the anion of a non-toxic pharmaceutically acceptable acid, [D] is a centrally acting drug species and [QC].sup.+ is the hydrophilic, positively charged ionic pyridinium salt form of a dihydropyridine.revreaction.pyridinium salt redox carrier.
Various aspects of the redox carrier system have been described in detail in Bodor U.S. Pat. No. 4,479,932, issued Oct. 30, 1984; Bodor U.S. Pat. No. 4,540,564, issued Sep. 10, 1985; Bodor et al U.S. Pat. No. 4,617,298, issued Oct. 14, 1986; UNIVERSITY OF FLORIDA's International Application No. PCT/US83/00725, published under International Publication No. WO83/03968 on Nov. 24, 1983; Bodor U.S. Pat. No. 4,727,079, issued Feb. 23, 1988; Bodor U.S. Pat. No. 4,824,850, issued Apr. 25, 1989; Bodor U.S. Pat. No. 4,829,070, issued May 9, 1989; Anderson et al U.S. Pat. No. 4,863,911, issued Sep. 5, 1989; Bodor U.S. Pat. No. 4,880,816, issued Nov. 14, 1989; Bodor U.S. Pat. No. 4,880,921, issued Nov. 14, 1989; Bodor U.S. Pat. No. 4,900,837, issued Feb. 13, 1990; UNIVERSITY OF FLORIDA's European Patent Application No. 88312016.4, published under European Publication No. 0327766 on Aug. 16, 1989; UNIVERSITY OF FLORIDA' s European Patent Application No. 89302719.3, published under European Publication No. 0335545 on Oct. 4, 1989; and numerous related publications. Among the redox carrier-drugs provided by the earlier chemical delivery system are dihydropyridine/pyridinium salt derivatives of dopamine, testosterone, phenytoin, GABA, valproic acid, tyrosine, methicillin, oxacillin, benzylpenicillin, cloxacillin, dicloxacillin, desipramine, acyclovir, trifluorothymidine, zidovudine, hydroxy-CCNU, chlorambucil, tryptamine, dexamethasone, hydrocortisone, ethinyl estradiol, norethindrone, estradiol, ethisterone, norgestrel, estrone, estradiol 3-methyl ether, estradiol benzoate, norethynodrel, mestranol, indomethacin, naproxen, FENU, HENU, 5-FU and many others.
The dihydropyridine redox carrier system has achieved remarkable success in targeting drugs to the brain in laboratory tests. Unfortunately, the dihydropyridine-containing derivatives suffer from stability problems, since even in the dry state they are very sensitive to oxidation as well as to water addition. Such problems have significantly complicated attempts to commercialize the system. Thus, a different carrier approach to brain-targeted drug delivery which would not include the inherently unstable dihydropyridine system would be desirable.
A phosphonate derivative of the antiviral agent DHPG has been described previously by Prisbe et al, J. Med. Chem. 1986, 29, 671-675. That compound, in which a ##STR3## is directly attached via to phosphorus-carbon bond to the antiviral drug, is structurally distinct from the phosphonate esters to which the present invention relates. Prisbe et al's phosphonate, unlike DHPG, was not active against herpes simplex virus types 1 and 2; however, it was reported to show moderate activity against HCMV in tissue culture.